Apodex 1.1 Launches: Beats DeepSeek V4 Pro in Agentic Benchmarks
ArtificialAnlys · x · 2026-08-31
Apodex has launched model 1.1, scoring 44 on the Artificial Analysis Intelligence Index, placing it alongside Kimi K2.6 and MiniMax-M3. The model excels in agentic tasks, achieving an Elo of 1348 on the GDPval-AA v2 benchmark, outperforming DeepSeek V4 Pro and Qwen3.7 Max. It also scored 70% on TerminalBench v2.1. However, it shows trade-offs in knowledge reliability (-21.9 AA-Omniscience) and verbosity (avg 17k output tokens). Priced at $0.05 per task, it offers attractive value compared to peers.
More from Models
- Diffusion LLMs Revise Mistakes and Speed Up Generation 5-10x — zainhas · 2026-08-31
- LLMs don't understand 'comprehensive': user finds AI search always misses data — ivan_bezdomny · 2026-08-31
- User Observation: Claude Seems to Particularly Enjoy Writing Firmware — _Stocko_ · 2026-08-31
- Darkbloom adds Qwen 3 VL 30B A3B model support — gajesh · 2026-08-31
- Speculation on Permanent Sandbagging by OpenAI and Anthropic — scaling01 · 2026-08-31
- Qwen3.8-Flash-Next Launches in INT4 and MXFP4; Auto-Round Tool Updated — HaihaoShen · 2026-08-31